The application of image analysis techniques to forward looking terrain database integrity monitoring

Abstract

This paper describes image analysis techniques that can be used for the comparison of terrain profiles obtained from a weather radar (WxR) in ground-mapping mode and the terrain information stored in a database. A synthetic vision system (SVS) provides the pilot with a virtual display of the surrounding environment to improve the situational awareness and reduce the number of aviation accidents caused by controlled flight into terrain. If the databases used for generating an SVS display do not have a quantifiable level of integrity, an integrity monitor may be a critical subsystem whose inclusion may guarantee the required integrity level in the terrain databases. The terrain information available for processing during the entire flight path is the stored terrain database and data obtained from onboard sensors. Integrity monitoring (IM) is done by making a comparison of these available data. The onboard sensor used for analysis in this research is a forward looking sensor, WxR. The comparison of the WxR data and the terrain database requires a common domain representation. This paper describes the generation of a common domain image representation from the WxR data and the terrain database and possible image analysis techniques to compare these images. The main feature used for the comparison of the images is the shadow region. The algorithms are tested using data obtained from a series of flight tests, one of which was conducted at Juneau, Alaska by Ohio University. The similarity results obtained are used as a criterion to determine any horizontal bias existent in the database.